Image Clustering
Models that partition the dataset into semantically meaningful clusters without having access to the ground truth labels.
Image credit: ImageNet clustering results of SCAN: Learning to Classify Images without Labels (ECCV 2020)
Papers
Showing 1–10 of 236 papers
All datasetsCIFAR-10CIFAR-100STL-10Imagenet-dog-15ImageNet-10MNIST-fullUSPSTiny ImageNetFashion-MNISTImageNetMNIST-testcoil-100
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | TURTLE (CLIP + DINOv2) | Accuracy | 72.9 | — | Unverified |
| 2 | MIM-Refiner (D2V2-ViT-H/14) | Accuracy | 67.3 | — | Unverified |
| 3 | SeLa | NMI | 66.4 | — | Unverified |
| 4 | PRO-DSC | Accuracy | 65 | — | Unverified |
| 5 | MIM-Refiner (MAE-ViT-H/14) | Accuracy | 64.6 | — | Unverified |
| 6 | TEMI MSN (ViT-L) | Accuracy | 61.6 | — | Unverified |
| 7 | TEMI DINO (ViT-B) | Accuracy | 58 | — | Unverified |
| 8 | MAE-CT (ViT-H/16 best) | Accuracy | 58 | — | Unverified |
| 9 | MAE-CT (ViT-H/16 mean) | Accuracy | 57.1 | — | Unverified |
| 10 | SeCu | Accuracy | 53.5 | — | Unverified |